Super Tilt Bro. for NES is a NES homebrew, featuring a fighting gameplay largely inspired by the Super Smash Bros. series.Status: nearing end of betaPlatforms: NES, Famicom, NES emulatorPrice tag: $0 (ROM download), estimated $30 NES cartridgeFeatures:Intense 2 players versus fighting Can run on actual NES and Famicom Free NES ROM download to play in an emulator Artificial intelligence for 1-player plays 4 stages to select from Please roast my game:As the game is nearing the end of beta, it would be interesting to know how good/bad it is from external point of view. And if there are big flaws that I am not aware of. Feedback can also help figure a target price tag for a version released on cartridge.Here some things to keep an eye on while testing:Are the controls easy to get? Are the controls pleasant? Are the stages good (can you rate it from favorite to worst)? Are graphics acceptable for a NES game? Anything that you would like to change or keep as it is now. Some things I'd like to know about your play session:Were you playing versus AI, versus a friend or both? Were you playing with a keyboard or some retro-gamepad? Did you played Smash Bros or any fangame before? Are you regularly playing actual 8-16bits games? (retro or homebrew) Note that the version 1.0 is nearing end of beta, but I actually plan on working on a 2.0 after that (with an improved engine). So do not hesitate to talk about big changes, there may be good ideas for the future :)
The game is inherently hard to get right. It is about programing a neural network and not every body have even the slightlest idea on how neural networks work. The game has the potential to be fascinating, has tons of people love neural networks and would love to understand it.
You may make it more accessible by creating an incremental "campaign", begining with a little brain almost configured and giving some instruction between scenarios. Also, the big text on the itch page, while fun to read at the begining is somewhat hard to keep focused on. Talking about charlie does not help, I understood it as "you are charlie, the man in front of a PICO-8" not as "There is a little alien walking on screen named charlie". Maybe you can keep this text but also add a more straight to the point part.
Great job on the amospheric part, it feels exactly what it should as a relaxing game.
Three points I think may be improved:
1. The levels gives too little clues about the soultion. When we run out of ideas, we are simply stuck.
2. The progression is shown nowhere. I would have loved to tell you on which level I was stuck, but I do not know :s
3. The music is on the point, but the loop is too short. For a game that lets us relax, this annoying little loop comes reguarily tease us.
I tested the version on scira.com, with a keyboard.
It is a good game, but from what I see everything is just one step from being great.
Graphics are awesome, except for the text in game master's speech-bubbles which is like if a blurring algorithm passed on it with not enough pixels to work. Also animations are bellow graphics quality, it is particularilly visible when the big fist destroys the castle, it just doesn't work (we understand the action, but not apreciate it).
Having different levels is cool, each harder than the previous. That's wonderfully done, but,... If next levels are harder, we make more points in the first ones, so the optimal strategy is to farm the level 1.
So, good game, push everything just a little forward and it could be really cool.
Plays nicely. Without adding to mechanics, polishing it the max you can could make it really enjoyable one-shot game. To do this way, here is what I think can be added:
(note: specific ideas are just suggestions)
** maybe no music as it will rapidly be boring
** but lots of sound effect: ambiant town sounds, car engine when moving, "click" when a photo is took, sounds associated to prend events,...
* Scoreboard (local and online)
* Visual clues
** How many money do events scores. Reading a number is boring, make the number bigger, blinking, jumping or whatever to denote visually that it is a good amount
** Flash the screen around the event when a photo is take
** Make the screen blinking like heartbeat or something to stress the player when the timer is almost terminated
* Less linear speed of the main triangle/character/photographer
** Have acceleration and decelerations
** Actually turn in intersections instead of going to the middle of the road then just be in the other direction
* Have more events at the same time
** Currently, we can catch almost all events, force the player to choose which to take and which to let go (adding strategy to the game)
** You maybe will need to have an intelligent placing of events, often grouping high revenues one, so that taking a low revenu one by greed puts you far away of most interesting events (all this while making greed tempting)
After that, adding more mechanics and tradeoff choices, like lulu1004 said, could make it more replayable.
Bug report: at the end of a game, if we wait some seconds, the "retry" button scrolls down and the game is stuck. I had to reload your itch.io page multiple times to study your game, a real player will not be so dedicated =)